{"title":"线性回归无分布置信椭圆的有限样本分析","authors":"Szabolcs Szentpéteri, Balázs Csanád Csáji","doi":"arxiv-2409.08801","DOIUrl":null,"url":null,"abstract":"The least squares (LS) estimate is the archetypical solution of linear\nregression problems. The asymptotic Gaussianity of the scaled LS error is often\nused to construct approximate confidence ellipsoids around the LS estimate,\nhowever, for finite samples these ellipsoids do not come with strict\nguarantees, unless some strong assumptions are made on the noise distributions.\nThe paper studies the distribution-free Sign-Perturbed Sums (SPS) ellipsoidal\nouter approximation (EOA) algorithm which can construct non-asymptotically\nguaranteed confidence ellipsoids under mild assumptions, such as independent\nand symmetric noise terms. These ellipsoids have the same center and\norientation as the classical asymptotic ellipsoids, only their radii are\ndifferent, which radii can be computed by convex optimization. Here, we\nestablish high probability non-asymptotic upper bounds for the sizes of SPS\nouter ellipsoids for linear regression problems and show that the volumes of\nthese ellipsoids decrease at the optimal rate. Finally, the difference between\nour theoretical bounds and the empirical sizes of the regions are investigated\nexperimentally.","PeriodicalId":501034,"journal":{"name":"arXiv - EE - Signal Processing","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Finite Sample Analysis of Distribution-Free Confidence Ellipsoids for Linear Regression\",\"authors\":\"Szabolcs Szentpéteri, Balázs Csanád Csáji\",\"doi\":\"arxiv-2409.08801\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The least squares (LS) estimate is the archetypical solution of linear\\nregression problems. The asymptotic Gaussianity of the scaled LS error is often\\nused to construct approximate confidence ellipsoids around the LS estimate,\\nhowever, for finite samples these ellipsoids do not come with strict\\nguarantees, unless some strong assumptions are made on the noise distributions.\\nThe paper studies the distribution-free Sign-Perturbed Sums (SPS) ellipsoidal\\nouter approximation (EOA) algorithm which can construct non-asymptotically\\nguaranteed confidence ellipsoids under mild assumptions, such as independent\\nand symmetric noise terms. These ellipsoids have the same center and\\norientation as the classical asymptotic ellipsoids, only their radii are\\ndifferent, which radii can be computed by convex optimization. Here, we\\nestablish high probability non-asymptotic upper bounds for the sizes of SPS\\nouter ellipsoids for linear regression problems and show that the volumes of\\nthese ellipsoids decrease at the optimal rate. Finally, the difference between\\nour theoretical bounds and the empirical sizes of the regions are investigated\\nexperimentally.\",\"PeriodicalId\":501034,\"journal\":{\"name\":\"arXiv - EE - Signal Processing\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - EE - Signal Processing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.08801\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - EE - Signal Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.08801","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
最小二乘(LS)估计是线性回归问题的典型解决方案。缩放 LS 误差的渐近高斯性常被用来构建 LS 估计值周围的近似置信椭圆,然而,对于有限样本,除非对噪声分布做出一些强有力的假设,否则这些椭圆并不具有严格的保证。本文研究了无分布符号扰动求和(SPS)椭圆外近似(EOA)算法,该算法可以在温和的假设条件下(如独立和对称噪声项)构建非渐近保证置信椭圆。这些椭球的中心和方向与经典渐近椭球相同,只是半径不同,而半径可以通过凸优化计算出来。在这里,我们为线性回归问题的 SPS 外椭圆的大小建立了高概率的非渐近上限,并证明这些椭圆的体积以最佳速率减小。最后,我们通过实验研究了我们的理论边界与经验区域大小之间的差异。
Finite Sample Analysis of Distribution-Free Confidence Ellipsoids for Linear Regression
The least squares (LS) estimate is the archetypical solution of linear
regression problems. The asymptotic Gaussianity of the scaled LS error is often
used to construct approximate confidence ellipsoids around the LS estimate,
however, for finite samples these ellipsoids do not come with strict
guarantees, unless some strong assumptions are made on the noise distributions.
The paper studies the distribution-free Sign-Perturbed Sums (SPS) ellipsoidal
outer approximation (EOA) algorithm which can construct non-asymptotically
guaranteed confidence ellipsoids under mild assumptions, such as independent
and symmetric noise terms. These ellipsoids have the same center and
orientation as the classical asymptotic ellipsoids, only their radii are
different, which radii can be computed by convex optimization. Here, we
establish high probability non-asymptotic upper bounds for the sizes of SPS
outer ellipsoids for linear regression problems and show that the volumes of
these ellipsoids decrease at the optimal rate. Finally, the difference between
our theoretical bounds and the empirical sizes of the regions are investigated
experimentally.